The Scanning Probe Microscope (SPM) is a powerful instrument in the nanometer scale science and technology. Among the many variations of SPM, the Atomic Force Microscope (AFM) is the most widely used and the most fundamental version. One prior art AFM is described in an article by G. Binnig, C. Quate, and Ch. Gerber in Phys. Rev. Lett. 56, 930 (1986). AFM has evolved since then, refining its capabilities and conveniences. In a commonly used configuration, a prior art AFM has a micro-machined cantilever with a sharp tip on its edge, and the AFM scans the sample or the cantilever with a piezoelectric tube. The deflection of the cantilever is measured by the AFM casting a laser beam on the cantilever and detecting the reflected beam with a position sensitive photo detector (PSPD). (See the article by G. Meyer and N. M. Amer, in Appl. Phys. Lett. 53, 2400 (1988).)
In such a configuration, the AFM has a high vertical sensitivity and is relatively easy to implement. In order to adjust the incident laser beam to fall on the cantilever and make the reflected beam hit the center of PSPD, an aligning mechanism with fine screws is used. A probing unit, including such an aligning mechanism plus the laser, PSPD, and cantilever, has considerable mass and it is difficult for the AFM to scan the probing unit at sufficiently high speed while maintaining accuracy. In the prior art, typical x scan speed is in the range of 0.1 Hz˜4 Hz and necessary z servo bandwidth is 100 Hz˜1 kHz. Such scan speed is acceptable but not satisfactory for the reasons discussed below. Therefore in certain conventional AFMs, the probing unit was kept stationary and the sample was scanned along x, y and z axes. See U.S. Pat. No. 5,157,251 granted to Albrecht et al. and U.S. Pat. No. 5,237,859 granted to Elings et al., both of which are incorporated by reference herein in their entirety.
However, large samples, such as large silicon wafers, cannot be scanned fast enough e.g. 1 KHz in z direction for sufficient vertical servo frequency response. For a description of this problem, see for example, U.S. Pat. No. 5,463,897 column 2, line 19-24. See also an article by P. K. Hansma, B. Drake, D. Grigg, C. B. Prater, F. Yashar, G. Gurley, and V. Elings, S. Feinstein, and R. Lal, J. Appl. Phys. 76, 796 (1994). The cantilever may be scanned while ensuring that the laser beam follows the cantilever motion to solve this problem. A simple method is miniaturizing the aligning mechanism, and scanning the whole probing unit. Such a scanning probe is implemented in the “AutoProbe M5” scanning probe microscope available from TM Microscopes, Veeco Metrology Group, 1171 Borregas Avenue, Sunnyvale, Calif. 94089, USA and described on the Internet at www.tmmicro.com.
However, such a miniaturized probing unit still has a considerable mass and it degrades the z servo response. It is also inconvenient to align the laser beam with tiny screws and a special tool had to be used. Another method is attaching lenses on the tube scanner such that the laser beam follows the cantilever motion and the reflected beam hits the same point on the PSPD. Such a tube scanner is implemented in the “Dimension 3100” microscope available from Digital Instruments, Veeco Metrology Group, 112 Robin Hill Road, Santa Barbara, Calif. 93117 and described on the Internet at www.di.com. See also U.S. Pat. No. 5,463,897 granted to Prater et al. which is incorporated by reference herein in its entirety. See also the article by P. K. Hansma, B. Drake, D. Grigg, C. B. Prater, F. Yashar, G. Gurley, and V. Elings, S. Feinstein, and R. Lal, J. Appl. Phys. 76, 796 (1994). However in this method, the laser beam does not perfectly follow the cantilever and the reflected beam does not remain on the exact same point on the PSPD, causing measurement errors and tracking force variations during x-y scan.
In addition, most AFMs have the common problems of scanning errors and slow scanning speed. A piezoelectric tube-based scanner commonly used in the prior art is not an orthogonal 3-dimensional actuator that can be moved in any of the three dimensions x, y and z independent of one another. Since the x-y motion relies on the bending of the tube, there is non-linearity and cross talk between x-y and z axes. AFMs can use position sensors to correct the intrinsic non-linearity of the piezoelectric tube as described in U.S. Pat. No. 5,210,410 granted to Barrett, and incorporated by reference herein in its entirety; see also an article by R. Barrett in Rev. Sci. Instrum. 62, 1393 (1991). However, z cross talk from flexing the tube cannot be eliminated and it causes background curvature effect and measurement errors. Using a tripod scanner does not improve the non-linearity and cross talk problem much. Furthermore the piezoelectric tube-based scanner has low resonance frequency (typically below 1 kHz) and does not have high force to drive a conventional probing unit at high speed.
In order to improve the orthogonality of the scanner, U.S. Pat. Nos. 6,310,342, 6,057,546 and 5,854,487 all granted to Braunstein, et al. (each of which is incorporated by reference herein in its entirety) describe the prior art use of a flexure stage for x-y scanning. However, since the z scanner described by Braunstein et al. is attached to the x-y scanner, the z scanner cannot move faster than the resonance frequency of the x-y scanner, which is about 100 Hz.
As mentioned above, the scanning speed of AFM is important. The scanning speed of AFM is usually limited by the z servo frequency response as described below. The z scanner needs to follow the sample features with appropriate feedback controller. As the scan speed in x direction is increased, the z scanner has to move up and down faster, requiring higher bandwidth in z servo system. However, the vertical servo frequency response cannot be higher than the resonance frequency of the z scanning system. z-scanning system means the z scanner and the supporting structure for the z scanner plus whatever the z scanner has to move in z direction. The resonance frequency of the z scanning system is reduced as more mass is loaded on to the z scanner. If the z scanner has higher push-pull force, the resonance frequency is reduced less.
Typical AFM has a few hundred Hz bandwidth in z servo system. Let's consider the case of 512 Hz. If we want to take a 256×256 pixel image, we can scan 1 Hz in x direction. (forward 256 plus backward 256) Of course, we can scan faster if the sample is smooth so that there is not much height variation between adjacent data points. Since we need to collect 256 lines of data, it takes 256 seconds (about 4 min.) to finish one scan. 4 min is a long time, and it is important to increase the scan speed. Typically, x-y scan speed is not limited by the x-y scanner bandwidth but limited by the z servo frequency response.
In AFMs, it is necessary to replace the cantilever frequently. The micro-machined cantilever is attached on a small chip (2×4 mm), and it is difficult to handle even with a tweezers. In order to improve the handling, the cantilever chip was mounted on an aluminum plate in the prior art. See U.S. Pat. No. 5,376,790 granted to Linker et al.; See also TM Microscopes CP, M5. Such a prior art plate had three slots, whose angles are 120° apart. These slots make contact with three balls. A spring clip was used to hold the chip mount. In another prior art (see TM Microscopes Explorer; see also U.S. Pat. No. 5,319,960 granted to Gamble et al.), a magnet was used to hold the chip mount but the chip mount sit directly on the magnet, which does not ensure the same cantilever position after replacement. Young et al. (see U.S. Pat. No. 5,705,814) have used a complicated method to align the cantilever.
Furthermore, the AFM head needs to be removed from the AFM frame from time to time. For convenient mount and un-mount of the AFM head, a dovetail groove has been made on the AFM head and a dovetail rail has been attached on the frame in the prior art. See, for example, TM Microscopes, CP; Digital Instruments, Dimension 3100. Since the AFM head needs to be firmly mounted on the frame, a fastening screw was used to tighten the dovetail.